the Third Week of Advent
Click here to join the effort!
Bible Dictionaries
Text, Versions, and Languages of Ot
Hastings' Dictionary of the Bible
or TEXT, VERSIONS, AND LANGUAGES OF OT
1 . Languages of the OT. The OT, except certain small sections, was written in Hebrew , and it has been preserved in its original language. But Jeremiah 10:11 , Daniel 2:4 to Daniel 7:28 , Ezra 4:8 to Ezra 6:18; Ezra 7:12-26 are in Aramaic , though it is disputed in the case of Daniel 2:4 to Daniel 7:28 whether this was the original language, or that of an Aramaic version which has replaced a Hebrew original. Hebrew and Aramaic alike belong to the group of languages known as Semitic, of which Assyrian (or the language of the ancient Assyrians and Babylonians) and Arabic are also important members.
2. The Hebrew language: Character and History . Hebrew is closely allied to PhÅ“nician, to the language of the Moabites represented by Mesha’s inscription ( c [Note: circa, about.] . b.c. 800), and to the language spoken in Canaan before (as well as after) the Hebrew invasion, known in part from the Canaanite glosses in the Tell el-Amarna tablets ( c [Note: circa, about.] . b.c. 1400), in part from Canaanitish names contained in ancient monuments, as, for example, the list of places in Canaan recorded as among his conquests by Thothmes iii. ( c [Note: circa, about.] . b.c. 1600). It is held by some scholars that the conquering Israelites adopted the language of Canaan, having previously spoken a language more nearly akin to Arabic (so, e.g. , Hommel, AHT [Note: HT Ancient Hebrew Tradition.] 120, 218). From the time at least when they were once well settled in the country, Hebrew was alike the colloquial and the literary language, of the Israelites. Some difference, such as is usual, no doubt always existed between the colloquial and the literary language though our knowledge of the colloquial is only such as we can draw by inference from the literature. But there came a time when Hebrew ceased to be the colloquial language, being replaced by Aramaic , and survived only as a literary language. The disuse of Hebrew in favour of Aramaic cannot be precisely dated, and was probably enough gradual; according to 2 Kings 18:26 , in the time of Isaiah (8th cent. b.c.), Aramaic was unintelligible to the Jewish populace, but as a language of diplomacy was spoken by Assyrian and Jewish officials alike. Apparently as late as Nehemiah (5th cent. b.c.) the colloquial language of the Jews in Palestine was still Hebrew, called ‘Jewish’ ( Nehemiah 13:24 as in 2 Kings 18:26 ). In the first century a.d., as the few sayings of the popular language preserved in the NT (such as Talitha cumi ) prove, it was Aramaic. Between these two dates, and, as we may infer from the increasing influence of Aramaic on the later books of the OT, considerably nearer the earlier than the later date, the change was made. Long before Aramaic replaced Hebrew as the spoken language, it exercised an influence through the spoken on the written language such as is commonly exercised by the language of one neighbouring people on another, that is to say, Hebrew borrowed words from Aramaic, as English borrows words from French and French from English. The Northern Kingdom was first brought into closer proximity with Aramaic-speaking peoples, and later the Southern Kingdom; and Aramaisms have consequently been regarded as pointing to a northern, or to a relatively late, origin of the writings in which they occur. Certainly any large presence of Aramaisms, and in particular any conspicuous Aramaizing of the syntax, due to the influence on their writings of the language which the later writers commonly spoke, such as we find, for example, in Daniel and Ecclesiastes, points to a late date.
Other languages besides Aramaic contributed to the vocabulary of Hebrew: Assyrian , indirectly through the Canaanites from the earliest times to an extent not easily to be defined, and later directly; Persian , after the Persian conquest of Babylon in 538; Greek , after the time of Alexander (332 b.c.); and Latin , after the establishment of Roman suzerainty over Judæa in the first century b.c. Latin words are found in the Hebrew of the Mishna, but not in the OT; a few Greek words in the latest writings of the OT (particularly Daniel, about b.c. 167) and very many in the Mishna; Persian words in some of the post-exilic literature (Esther, Canticles, Tobit).
3. The Hebrew alphabet vowelless . The Hebrew alphabet used by the OT writers consisted of twenty-two consonants: it contained no vowels, in this resembling Phœnician, Moabitic, and the ancient Arabic and Syriac alphabets. Our knowledge of the pronunciation of Hebrew words, as far as the vowels are concerned, depends on three main sources: (1) Jewish tradition, which is embodied in vowel signs invented between the 4th and 9th centuries a.d., and written under, over, or in the consonants of the ancient text; (2) the Greek versions, which transliterate a large number of Hebrew words, especially, but by no means only, the proper names; (3) the Assyrian texts: these, being written in a language which expressed in writing vowel sounds as well as consonantal, give us the vowels of such Hebrew names as they cite.
Though in the oldest Hebrew MSS of the Bible the consonants of the original text are accompanied by the vowels which express at once the traditional pronunciation and the traditional interpretation of the text, it is now as generally accepted that the vowels formed no part of the original text as that the earth revolves round the sun. Down to the 17th century it was otherwise; and that century was marked by a final and keen discussion of this point.
4. Transliteration of Hebrew adopted in this article . Since considerable importance attaches to this Jewish tradition as to the pronunciation, it will be necessary to represent the vowels in our discussion of the text, but it is important also to indicate their secondary origin and subordinate position. Throughout this article, then, the Hebrew consonants will be represented by equivalent or approximately equivalent English capitals, except the 1st and 16th letters of the Hebrew alphabet, which, being gutturals with no approximate equivalent in English, will be retained in their Hebrew form ( ×¢ , × ), and may be passed over unpronounced by the English reader. The vowels will be represented by English small letters printed under the consonant after which they are to be pronounced; thus D aBaR, pronounced dabar . The Jewish scholars distinguished by different signs between long and short vowels; no attempt will be made here to mark these distinctions, and the peculiar half-vowels, the shÄ•va’s , as they are termed, will be left unrepresented. Letters doubled in pronunciation, but without a vowel between them, were represented by the letter written once, not twice. The Hebrew vocalists distinguished these doubled letters by inserting a dot in the middle of them. This dot or daghesh will be represented here by the sign | above the letter: thus DiBeR, pronounced dibber .
5. Date of the addition of vowels to the OT text . The date at which the vowels were attached to the consonants of the Hebrew text can be determined only within broad limits. It was after the beginning of the 5th cent. a.d., for the way in which Jerome speaks leaves no room for doubt that the Hebrew Scriptures in his day were un-vocalized; it must have been before the 10th cent., for the fully developed system is employed in the earliest Hebrew Biblical MSS, which date from the beginning of the 10th cent. (or, according to some, from the 9th cent.).
6. Earlier attempts to represent vowel sounds . Long before the invention of vowel points certain consonants had been used, though neither systematically nor consistently, to indicate the vowel sounds: thus H [Note: Law of Holiness.] was used to indicate a , and sometimes e; W to indicate o or u , Y to indicate i . This practice in some measure goes back to the times, and doubtless also to the actual usage, of some of the writers of the OT; but in many cases these consonants used to indicate vowels were added by scribes or editors. This we learn from the fact that passages which happen to occur twice in the OT differ in the extent to which, and the particular instances in which, these letters are employed. Psalms 18:1-50 occurs not only in the Psalter, but also in 2 Samuel 22:1-51; the Psalm expresses these consonants used vocalically 17 times where 2Sam. does not, e.g. 2Sam. writes ḲDMNY ( 2 Samuel 22:6 ) and HḤ ÅžYM ( 2 Samuel 22:31 ), where the Ps. writes ḲDMWNY and HḤWá¹¢YM. In some cases Rabbinic discussions prove that words now written with these vowel letters were once without them; so, e.g. , it appears from a discussion attributed to two Rabbis of the 2nd cent. a.d. that in Isaiah 51:4 the word L × WMY (‘my nation’ RV [Note: Revised Version.] ) was at that time written without the W, thus L × MY. The importance of this fact for the textual criticism will appear later.
7. Character of evidence for the text of OT . The text of the OT has been transmitted to us through circumstances singularly different from those which mark the transmission of the NT text; and the results are a difference in the relative value attaching to different classes of evidence, and a much less close and sure approach to the original text when the best use has been made of the material at our disposal. Quotations play a much less immediate and conspicuous part in the criticism of the OT than in the criticism of the NT; and here we may confine our attention to the nature of the evidence for the text of the OT furnished by (1) Hebrew MSS, (2) ancient Versions.
8. (1) Hebrew MSS . One well-established result of the examination of Hebrew MSS is that all existing MSS are derived from a single edition prepared by Jewish scholars in accordance with a textual tradition which goes back substantially to the 2nd cent. a.d., but became increasingly minute. This is proved by the existence in all MSS of the same peculiarities, such as the occurrence at certain places of letters smaller or larger than the normal, of dots over certain letters, or broken or inverted letters. For example, the H in the word BhBR × M ( Genesis 2:4 ) is written small in all Hebrew MSS; it was doubtless written originally so by accident or owing to pressure of room; but under the influence of a school of Jewish scholars, of whom R. Aqiba in the 2nd cent. b.c. was a leading spirit, all such minutiae of the Scripture acquired a mystic significance. Thus the word just cited really means ‘when they were created,’ but the small H was taken to mean that the words were to be translated ‘in the letter H he ( i.e. God) created them’ (the heavens and the earth), and this in turn led to much curious speculation. As another illustration of this method of Interpretation, which was so important in securing from the 1st or 2nd cent. a.d. onwards a remarkably accurate transmission of the text, the case of the word WYYẒR In Genesis 2:7 may be cited. The word means ‘And he formed’; an alternative orthography for the word is WYẒR (with one Y). Why, it was asked, was it here written with two Y’s? Because, it was answered, God created man with two YẒRS ( i.e. two natures), the good nature and the bad. In order to secure the perpetuation of the text exactly as it existed, a mass of elaborate rules and calculations was gradually established; for example, the number of occurrences of cases of peculiar orthography, the number of words in the several books, the middle word in each book, and so forth, were calculated and ultimately embodied in notes on the margins of the MSS containing the Scriptures. This textual tradition is known as the Massorah , and those who perpetuated it as Massoretes . The Massorah also Includes a certain number of variant or conjectural readings; In this case the one reading ( Kethibh ‘written’) stands in the text, but provided with vowels that do not belong to the consonants in the text, but to the consonants of the alternative reading ( Qerç ‘read’) given in the margin. E.g. , in Job 9:30 the word BMW, which means ‘with,’ should, if vocalized, have the vowel o over the W; but in the Hebrew text the vowel actually supplied to the word is e under the M, which is the vowel that really belongs to the marginal reading BMY, and this means ‘in the water of.’ These Massoretic variants are for the most part relatively uninteresting. The value of the Massorah in perpetuating a form of the Hebrew text for many centuries has doubtless been great; but it has also long served to obscure the fact that the text which it has perpetuated with such slight variation or mutilation was already removed by many centuries from the original text and had suffered considerably.
In spite of the Massorah, certain minute variations have crept into the Hebrew MSS and even into the consonantal text. The vowels, it must be repeated, are merely an interpretation of the original text of Scripture, and not part of it, and different Hebrew MSS show as a matter of fact two distinct systems of vocalization, with different symbols.
9. The earliest MSS . Among the earliest Hebrew Biblical MSS are the Prophetarum posteriorum codex Babylonicus Petropolitanus , dated a.d. 916; a codex of the Former and Latter Prophets now in the Karaite synagogue at Cairo, and written, if correctly dated, in a.d. 895; a codex of the entire Bible, written by Samuel ben Jacob, now at St. Petersburg, and written, if the dating be genuine, in a.d. 1009.
10. Critical editions of the Massoretic text . The most accurate reproductions of the Massoretic text are the edition of the Hebrew Bible by S. Baer and Fr. Delitzsch and that by C. D. Ginsburg. These are critical editions of the Massoretic text, but make no attempt to be critical editions of the OT text, i.e. they make no use whatever of the Versions or of any other evidence than the Massoretic tradition.
11. The Samaritan Pentateuch . Before passing from the evidence of Hebrew MSS we have to note that for the Pentateuch, though unfortunately for the Pentateuch only, we have the invaluable assistance of a Hebrew text representing an entirely different recension. This is the Samaritan Pentateuch . The Samaritan Pentateuch is a form of the Hebrew text which has been perpetuated by the Samaritans. It is written in the Samaritan character, which far more closely resembles the ancient Hebrew characters than the square Hebrew characters in which the Massoretic MSS are written, and is without vowels . The available MSS of the Samaritan Pentateuch are considerably later than the earliest Massoretic MSS; nor is it probable that the copy at Nâblus, though perhaps the earliest Samaritan MS in existence, is earlier than the 12th or 13th cent. a.d. But the value of the recension lies in the fact that it has descended since the 4th cent. b.c. in a different circle, and under different circumstances, from those which have influenced the Massoretic MSS. Though in some respects, as for example through expansion by insertion of matter from parallel passages, the Samaritan is more remote than the Jewish from the original text, it has also preserved better readings, often in agreement with the LXX [Note: Septuagint.] . An instance is Genesis 4:8; here in the ordinary Hebrew MSS some words spoken by Cain have certainly dropped out; the fact is obscured in the RV [Note: Revised Version.] (text), which mistranslates; the Hebrew text really reads, ‘And Cain said to Abel his brother’; the Samaritan text and the LXX [Note: Septuagint.] have the additional words, ‘Let us go into the field’; this is probably right (see next clause).
12. The Samaritan Targum . No thoroughly critical edition of the Samaritan Pentateuch at present exists. The material for establishing a critical text consists of the several MSS and also of the Samaritan Targum a translation of the Samaritan recension into an Aramaic dialect. The colloquial language of the Samaritans, like that of the later Jews, was different from that in which the Scripture was written.
13. Papyrus fragment of OT text . Thanks to a recent discovery, we have a further witness to a fragment of the Hebrew text of the Pentateuch. This is the Nash papyrus. The papyrus is apparently not later than the 2nd cent. a.d.; and it contains the Ten Commandments and Deuteronomy 6:4 f. in Hebrew. The text, which is of course unvocalized, is several times in agreement with the LXX [Note: Septuagint.] against the Massoretic text. This fragment was edited by Mr. S. A. Cook in PSBA [Note: SBA Proceedings of Soc. of Bibl. Archeology.] (Jan. 1903).
14. (2) Versions: Earliest MSS . We come now to the second main branch of evidence for the text of the OT. The evidence of Versions is of exceptional importance in the case of the OT. In the first place, the actual MSS of the Versions are much older than the earliest Hebrew MSS; the earliest Hebrew MSS date from the 10th cent. a.d. but there are Greek MSS of the OT of the 4th cent. a.d. and there is a Syriac MS of the greater part of the Pentateuch of the date a.d. 464. But secondly, and of even greater importance, the Versions, and especially the LXX [Note: Septuagint.] , represent different lines of tradition; in so far as the original text of the LXX [Note: Septuagint.] itself can be established, it is a witness to the state of the text some two to four centuries before the date at which the stereotyping of the Hebrew text by the Massoretes took place.
The Versions of the OT are either primary, i.e. made direct from the Hebrew text, or secondary, i.e. made from a Version. Secondary Versions are of immediate importance in establishing the true text of the primary version from which they are made; and only indirectly witness to the Hebrew text. Among them the Old Latin Version is of exceptional importance in determining the text of the LXX [Note: Septuagint.] . On this and other versions of the LXX [Note: Septuagint.] , see Greek Versions of OT, § 11 .
15. Brief account of the Primary Versions . The Primary Versions of the OT, arranged in (approximately) chronological order, are as follows:
(1) The earliest Greek Version , commonly known as the Septuagint . The earliest part of this version, namely, the translation of the Pentateuch, goes back to the 3rd century b.c. The remaining parts of the OT were translated at different later periods; but the version was probably, in the main at least, complete before the end of the 2nd cent. b.c. See Gr. Versions of OT.
(2) The Targums . These Aramaic versions may be considered next, inasmuch as they rest on a tradition earlier than the date of the versions yet to be mentioned; it is probable, however, that no Targum was actually committed to writing till some centuries later, after the later Greek versions, perhaps, too, after the Syriac Version, had been made.
The quotation from Psalms 22:1 in Matthew 27:46 || Mark 15:34 is in Aramaic; and Ephesians 4:8 agrees more closely with the Targum than with the Hebrew text of Psalms 68:4 . From these facts we may perhaps infer that an Aramaic version bad to some extent become orally fixed by the 1st cent. a.d.
The Targumsarein large part very free, and even diffuse, paraphrases rather than translations of the Hebrew text. They owe their origin to the custom of explaining the Hebrew passages of Scripture read in the synagogues in the language spoken by the people, which was Aramaic. The earliest (as is most generally believed) and least paraphrastic of these versions is the Targum of Onkelos on the Pentateuch; it does not appear to have been committed to writing before the 5th cent. a.d., and is first mentioned by name by Saadiah Gaon in the 9th century. Far more paraphrastic is the Targum of the Pentateuch known as the Targum of Jonathan , or the Jerusalem Targum . Fragments of yet a third Targum of the Pentateuch survive, and are known as the 2nd Jerusalem Targum . Quite distinct from these is the Samaritan Targum , which is a translation of the Samaritan recension of the Hebrew text (see § 11 ). The chief Targum of the Prophets is that known as the Targum of Jonathan ben Uzziel : it is not much younger than the Targum of Onkelos, and is by some considered to be even earlier. There are also fragments of another Targum of the Prophets. Targums of the Hagiographa (with the exception of Ezra, Nehemiah, and Daniel) exist, and there are two of the Book of Esther. Cf. art. Targume.
The text of the Targums will be found in Walton’s (and other) polyglots, with a Latin translation. Onkelos has been separately edited by Berliner (1884), and the Prophets and Hagiographa by Lagarde (1872, 1874). See, further, Hastings’ DB [Note: Dictionary of the Bible.] , art. ‘Targum.’ There is an English translation of the Targums of the Pentateuch by Etheridge (2 vols., London, 1862 1865).
(3), (4), and (5) The Greek Versions (which have survived in fragments only) of Aquila, Theodotion, and Symmachus, all of the 2nd cent. a.d. See Greek Versions of OT, §§ 15 18 .
(6) The Syriac Version , commonly called the Peshitta. The date at which this version was made is unknown. The earliest extant MS of part of this version is, as stated above, of the year 464 a.d.; and the quotations of Aphraates (4th cent. a.d.) from all parts of the OT agree with the Peshitta. The character of the version differs in different books, being literal in the Pentateuch and Job, paraphrastic for example in Chronicles and Ruth. The text in the main agrees closely with the Massoretic Hebrew text, though in parts ( e.g. in Genesis, Isaiah, the Minor Prophets, and Psalms) it has been influenced by the LXX [Note: Septuagint.] .
(7) The Vulgate . The Old Latin Version was a translation of the LXX [Note: Septuagint.] . To Christian scholars acquainted with Hebrew the wide differences between the LXX [Note: Septuagint.] and versions derived from it and the Hebrew text then current became obvious. As it seemed suitable to Origen to correct the current LXX [Note: Septuagint.] text so that it should agree more closely with the Hebrew, so at the close of the 4th century Jerome, after first revising the Old Latin, making alterations only when the sense absolutely demanded it, prepared an entirely fresh translation direct from the Hebrew text. The Vulgate is derived from this direct translation of Jerome’s from the Hebrew in the case of all the canonical books of the OT except the Psalms; the Psalms appear commonly in editions of the Vulgate in the form of the so-called Gallican Psalter; this was a second version of the Old Latin, in which, however, after the manner of Origen’s Hexaplaric text, the translation was brought nearer to the current Hebrew text by including matter contained in the later Greek versions but absent from the LXX [Note: Septuagint.] , and obellzing matter in the LXX [Note: Septuagint.] which was absent from the later versions. Jerome’s Latin version of the Psalms, made direct from the Hebrew, has been edited by Lagarde ( Psalterium juxta HebrÅ“os Hieronymi , 1874). On the extent to which editions of the Vulgate differ from Jerome’s translation, see Vulgate. In some cases additional matter ( e.g. 1 Samuel 14:41 , on which passage see § 24 ) has been incorporated from the Old Latin.
The effect of the substitution of Jerome’s version from the Hebrew text for the Old Latin version of the LXX [Note: Septuagint.] was to give the Church a Bible which was more elegant and intelligible and in much closer agreement with the Hebrew text current in the 4th cent. a.d., but which at the same time was in many passages more remote from the original text of the OT.
16. Two groups of versions. Pre-eminence of the Septuagint . Judged from the standpoint of their importance for recovering the original text of the OT, and for the kind of service which they render to OT textual criticism, the primary versions fall into two groups: (1) the LXX [Note: Septuagint.] , (2) the rest. The LXX [Note: Septuagint.] differs, and often differs widely, from the Massoretic text; the remaining versions closely agree with it: the LXX [Note: Septuagint.] dates from before the Christian era and, what is more significant, from before the rise of the Massoretic schools; the remaining versions date from after the Christian era, and, with the possible exception of the Syriac, from after the close of 1st cent. a.d. The agreement of these versions made direct from the Hebrew text at various dates subsequent to 100 b.c. confirms the conclusion suggested above, that since that date the Hebrew text has suffered relatively little in course of transmission. Such variations as do occur in these versions from the Hebrew consist largely (though not exclusively) of variations in the Interpretation of the consonants, i.e. while presupposing the same consonants as the present Hebrew text, they presuppose also that these consonants were pronounced with other vowels than those which were added to the text after the 5th cent. a.d. These variations therefore do not, strictly speaking, represent variants in the text of the OT, but merely in the commentary on that text, which at the time the versions were made was still oral, and only later was committed to writing in the form of vowels attached to the consonants, of which alone the Scripture proper consisted.
A fuller discussion of the versions of the OT other than the LXX [Note: Septuagint.] would carry us into minutiœ of the subject which do not belong to a brief sketch such as the present. On the other hand, the LXX [Note: Septuagint.] claims further attention even here.
17. The early history of the Hebrew text . The history of the Hebrew text since the 2nd cent. a.d. is uneventful; it is a history of careful transmission which has preserved the text from any serious deterioration since that date. But the fortunes of the text before that date had been more varied and far less happy. They cannot be followed completely, nor always with certainty. But the main fact is abundantly clear, that between the ages of their several authors and the 2nd cent. a.d. the Hebrew Scriptures had suffered corruption, and not Infrequently very serious corruption. Nor is this surprising when it is remembered that the text in that period consisted of consonants only, that in the course of it the character of the writing was changed from the Old Hebrew to the square character still in use (the difference between the two being greater than that between old black letter type and the Roman type now commonly used), that in the earlier part of the period copies of the books cannot have been numerous, and that in times of persecution copies were hunted for and destroyed ( 1Ma 1:56 f.) We are here concerned, of course, merely with such changes as crept into the text accidentally, or such minor changes as the introduction of the expressed for the implicit subject, which belong to the province of textual criticism. The larger changes due to the editing and redacting or union of material belong to the province of higher criticism, though in the case of the OT it is particularly true that at times the line between the two is not sharply defined. Our chief clues to the earlier history of the Hebrew text, and to the solution of the problems connected with it, will be found in a comparison of the Hebrew text with the Septuagint version, and in certain features of the Hebrew text itself. The remainder of this article will be devoted to elucidating and illustrating these two points.
18. The Hebrew Text between c [Note: circa, about.] . b.c. 250 and c [Note: circa, about.] . a.d. 100. The LXX [Note: Septuagint.] and the Massoretic Text . The materials for forming a judgment on the general character of the changes undergone during this period by the Hebrew text, and for the existence of early variant readings in particular passages, are to be drawn mainly from a comparison of the LXX [Note: Septuagint.] with the Hebrew texts. A much smaller amount of material is to be derived from the quotations in the NT and other early Jewish works, such as the Book of Jubilees, written, according to Dr. Charles, at the close of the 2nd century b.c.; but so far as it goes this material bears witness of the same general character as that of the LXX [Note: Septuagint.] .
19. A correct solution of the main problem here raised depends on three things: (1) the establishment of the original text of the LXX [Note: Septuagint.]; (2) the detection of the Hebrew text which lay before the translators; and (3) In cases where the Hebrew text there recorded differs from the present Hebrew text, the determination of the more original of the variants. A complete solution of the problems will never be reached, for it will be no more possible to establish beyond dispute the original text of the LXX [Note: Septuagint.] than the text of the NT; the detection of the underlying Hebrew text must inevitably often remain doubtful; and when variants are established, there will be in many cases room for differences of opinion as to their relative value. But though no complete solution is to be hoped for, a far greater approximation to such a solution than has yet been reached is possible. A good beginning (though no more) towards the recovery of the original text of the LXX [Note: Septuagint.] has been made (see Greek Versions of OT, § 13), but of really systematic work on the recovery of the underlying Hebrew text there has been far too little. What commonly happens is that in particular passages where the sense of the LXX [Note: Septuagint.] and of the Hebrew text differs, the Greek is re-translated without exhaustive reference to the methods of the translators, and the re-translation thus obtained is cited as the variant. In many cases the true variant even thus has undoubtedly been obtained, but in many others a closer and more systematic investigation of the methods and idiosyncrasies of the translators has shown or will show that, through misinterpretation, the support of the LXX [Note: Septuagint.] has been cited for variants which there is no reason for believing ever had any existence.
20. Distinction between real and apparent variants . A difference in sense between the Greek version and the Hebrew text as subsequently interpreted by no means necessarily points to a variation in the Hebrew text that underlay the version.
For example, parts of the three Hebrew verbs Å BH ( to lead captive ), and YÅ B ( to dwell ) and of Å WB ( to return ) are indistinguishable in the Hebrew consonantal text; the letters WYSB may have among others the following meanings, and he dwelt, and he returned, and he brought back, and he took captive .
The substitution of one of these meanings for the other occasionally reduces the Greek version to nonsense; inconvenient as this must have been for those who used that version, or versions, like the Old Latin, made from it, it presents no difficulty to those who are attempting to recover the Hebrew original of the Greek version. It may sound paradoxical, yet it is to a large extent true, that for textual criticism the LXX [Note: Septuagint.] is most useful when it makes least sense; for when a passage makes no sense in the Greek, but can be explained as a translation from the Hebrew, we have the best of reasons for believing that we have before us the original text of the Greek, and through it can recover a Hebrew text of early date. Copyists and translators do not deliberately turn sense into nonsense, and sense does not frequently, through mere accidents of transmission, become the particular form of nonsense that can be accounted for by a misunderstanding of a Hebrew original.
As a further illustration we may refer to the Greek translation of the letters BY; these very commonly occur with the meaning in me , but they also represent a particle of entreaty Oh! or I pray!; this particle occurs but rarely, about a dozen times altogether, and its existence was unknown to some of the Greek translators. In the Pentateuch and Joshua it is correctly rendered; but else where it is rendered ‘in me’ with ridiculous results, as the English reader will see if he substitutes these words for ‘Oh’ in Judges 6:13 , 1 Samuel 1:26 . But again, there is no difficulty in seeing beneath the nonsense of the Greek the true sense and the actual reading of the Hebrew. The ignorance of the translators is as useful to the textual critic as their knowledge.
21 . Euphemistic translations . But there are many variations in sense which point to no real textual variants, though both Hebrew and Greek in themselves yield a good sense.
The last clause of the 19th Psalm in the AV [Note: Authorized Version.] , ‘O Lord, my strength and my redeemer, ‘reads admirably; but though the translators give us no clue to the fact, it is not a translation of the Hebrew, it is a translation of the LXX [Note: Septuagint.] . The Hebrew reads ‘My rock and my redeemer’ (so RV [Note: Revised Version.] ). In this case the LXX [Note: Septuagint.] rendering is due not to ignorance, but to religious scruple; their rendering is a euphemism. So in Genesis 5:24 the Greek version substitutes ‘Enoch was well-pleasing to God’ (hence Hebrews 11:5 ) for the anthropomorphic walked with God’ of the Hebrew text; in these cases, if we had not also the Hebrew text we could not discover the original from the LXX [Note: Septuagint.] with certainty, or, perhaps, even be sure that the translators were paraphrasing and not translating.
22. Relative values of Greek version and Hebrew text . These illustrations may suffice to show both that much care is required in using the LXX [Note: Septuagint.] for the recovery of the Hebrew underlying it, and also that it is wide of the mark to depreciate the textual value of the version by emphasizing the ignorance of the translators. Before either the fullest or the securest use of the version can be made, an immense amount of work remains to be done; but the importance of doing this work is clear, for even the most cautious deductions have already proved that the text underlying the LXX [Note: Septuagint.] and the present Hebrew text differ widely, and that in many Instances the LXX [Note: Septuagint.] text is superior. The relative values differ in the case of different books; and to avoid misunderstanding it should be added that in no case would a simple translation of the LXX [Note: Septuagint.] bring us as near to the sense of the original document as a translation from the Hebrew text; nor would it be possible, unless the Hebrew text had survived, to detect by means of the LXX [Note: Septuagint.] the correct text and the sense of the original. Issues are sometimes confused, and the distinctive characteristics and virtues of our two chief witnesses to the text of the OT obscured, in discussions as to the relative values of the LXX [Note: Septuagint.] and the Massoretic text. Perhaps the most important general point to remember is that neither the one nor the other would be nearly as valuable by itself as it is when used in combination with the other.
23 . Examples of important readings preserved by the Greek Version only . We may now pass to some illustrations of important variations in which the LXX [Note: Septuagint.] has clearly preserved an earlier text than the Hebrew. These are much less numerous in the Pentateuch than elsewhere; probably the Law, as the most important Scripture, received at an early period something approaching to that great care in transmission which was later extended to the entire OT. It is the more remarkable, therefore, that in one section of the Pentateuch ( Exodus 35:1-35; Exodus 36:1-38; Exodus 37:1-29; Exodus 38:1-31; Exodus 39:1-43 ) we find striking differences in the arrangement of sections in the Hebrew and Greek texts. Other instances of different arrangement or of marked differences in the extent of the material occur in the Books of Job and Jeremiah (see, further, Swete, Introd. to the OT in Greek , 221 ff.). This type of difference connects the textual with the higher criticism of these books, and cannot be pursued further here.
24. In some cases matter subsequently lost (through homoioteleuton or otherwise), and now absent from the Hebrew text, survives in the Greek.
A striking illustration of this occurs in 1 Samuel 14:41 . The Hebrew text underlying the Greek version reads,’ Saul said unto Jahweh, the God of Israel [wherefore hast thou not answered thy servant to-day? If this iniquity be in me, or in Jonathan my son, O God of Israel, give Urim, but if this iniquity be in thy servant Israel], give Thummim.’ The words in square brackets are absent from the Hebrew text, but certainly belonged to the original, and the origin of the error is clear: the scribe’s eye accidentally passed from the first occurrence of ‘Israel’ to the third, and the intervening words were lost. With the loss of these the sense of the last two words ‘give Thummim’ became obscure, and the punctuators, followed by RV [Note: Revised Version.] , gave them an indefensible interpretation.
25. In other cases the Greek version is nearer to the original by its relative brevity; the additional matter now present in the Hebrew text was subsequently interpolated.
As an instance of this we may cite 1 Kings 6:20-21 , which RV [Note: Revised Version.] , following the Hebrew text, renders, ‘And he covered the altar [with cedar. So Solomon overlaid the house within with pure gold: and he drew chains of gold across] before the oracle; and he overlaid it with gold.’ The bracketed words are absent from the Greek; it is probable that of these words ‘with cedar’ stood in the original text, but that the rest were absent. The Greek text has also for the first four words above (before the bracket) the (superior) reading, ‘And he made an altar.’
26. At times, when either the sense or the text of both the Hebrew and the Greek is remote from the original, it is possible, from a comparison between the two, to recover the original.
An interesting example of this is furnished by Isaiah 37:27 f. = 2 Kings 19:26 f. RV [Note: Revised Version.] , following the Heb. text, renders, ‘They were as the grass of the field, and as corn (Is. ‘a field of corn’) blasted before it be grown up . But I know thy sitting down and thy going out and thy coming in.’ The Hebrew text of the underlined words is LPNY QMH WSBTK; the Hebrew equivalent of ‘I know’ stands much lower in the sentence, and though it may with difficulty be taken as in the RV [Note: Revised Version.] , more naturally demands a different object. A reading of the Greek text preserved only in a Syriac version of it, but nevertheless probably the original reading of the Greek text, has, for the same underlined words, ‘before thy rising up, and thy sitting down’; this presupposes the Hebrew LPNY QMK WSBTK, which differs from the present Hebrew text by one letter only. The Hebrew text here presupposed is probably original, but has been misunderstood by the translators. The first word, if vocalized as in the Hebrew text and by the Greek translators L iPNeY, means before , but if vocalized LaPNaY it means before me . Adopting the latter vocalization, we recover (at least so far as the three words are concerned) the original sense,’ They were as grass of the field … and as corn that is blasted. Before me is thine uprising and thy down sitting (cf. Psalms 139:2 ); and thy going out and thy coming in I know.’ So great is the difference in sense that the corruption of a single letter may make in a text which contained only consonants, and no marks of punctuation whatever. The true reading of the Hebrew in this case was first divined by Wellhausen; it remained for Mr. Burkitt to point out that it was the reading of the Greek translators.
27 . The Hebrew text before the date of the Greek version . If the Hebrew text suffered to a very considerable extent in the ways just illustrated, during the three or four centuries that intervened between the time when the LXX [Note: Septuagint.] version was made and the time when the Hebrew text was stereotyped and the later Greek versions were made, by nothing short of a stupendous miracle could the text have been preserved free from errors of transmission, during the centuries that separate the original autographs from the date of the Greek version. This intervening period differs, of course, widely in length; between the age of Isaiah and the Greek translation of the Book of Isaiah lay some six centuries; between the age of Deborah ( Judges 5:1-31 ) and the translation of Judges little short of a thousand years; between the age of David ( 2 Samuel 1:19 ff.) and the translation of Samuel 800 or 900 years. On the other hand, between the compilation of the Hexateuch, or the first composition of books such as Ecclesiastes or Daniel, and the translations in the several cases, not more than a couple of centuries elapsed.
28. Means of detecting early corruption of Hebrew text . Though the general fact that the present Hebrew text contains corruptions that date from these earlier centuries cannot reasonably be questioned, the detection of the actual cases of early corruption is necessarily difficult, and only within limits is it possible. We are obviously far worse situated in attempting to determine corruptions of this date than corruptions of later date; the LXX [Note: Septuagint.] often indicates the presence of the later corruptions, but we have no external clue to the earlier corruptions. We have to rely entirely on indications in the Hebrew text itself. One of these indications will of course be the occurrence of nonsense, for the original autographs were intended to convey an intelligible meaning. Another indication will be the occurrence of bad grammar unless in the case of a particular writer there is reason for supposing that he was not master of the language which he wrote. An interesting illustration of the way is which the latter indication may serve is furnished by some of the references to the ark.
The ark is called in Hebrew H × RN the ark , where the first letter is the Hebrew article; or × RN BRYT YHWH the ark of the covenant of the Lord; where a word in Hebrew is defined by a following genitive it cannot be preceded by the article, so in this second phrase we have × RN, not H × RN. Now, in certain passages ( e.g. Joshua 3:17 ), our present Hebrew text has the grammatically impossible combination H × RN BRYT YHWH; some corruption theo is present here; and it is probable that the original text had only H × RN the ark , and that the two following words are due to the intrusion ioto the text of an annotator’s explanation.
29 . Negative and positive judgments: the justification of conjectural emendation and its limitations . The ultimate task of textual criticism is to recover as far as possible the actual words of the original; an intermediate task of the textual criticism of the OT is to establish all the real variants of the Hebrew text underlying the Greek version, and in each case to determine the relative value of the variants. In this way the text which was the common source of the Greek translators and that of the Jewish scholars of the 2nd cent. a.d. is as far as possible recovered. So far negative and positive judgments must necessarily accompany one another; we say, Here the Hebrew text is right, and the Greek text Wrong, or vice versa . But when we have recovered that common source of the Hebrew and Greek texts, it is wise to distinguish sharply between negative and positive critical judgments. The general fact that there are early errors in the Hebrew text must, as we have seen, be admitted; and, further, no sound criticism of the Hebrew text can proceed far without being compelled to say, This or that is corrupt, even though the Greek version agrees with the Hebrew text or cannot be shown to have differed from it. In some cases where this negative judgment can be passed with confidence, it may be possible with scarcely less confidence to pass to the positive statement. These words are a corruption of these other words; that is to say, the text in such cases can be restored by conjecture; but in many cases where the first judgment These words are not the original text must be passed, the second judgment ought only to take the form It is possible that such and such words or something like them were in the original text. In brief, we can more often detect early corruption than restore the text which has been corrupted. The reason should be obvious. Nonsense (to take the extreme case) must be due to corruption, but the sense which it has obscured may altogether elude us, or, at best, we may be able to discern the general sense without determining the actual words.
There can be no question that it is nonsense to say, as the Hebrew text does, that Saul, who was anointed king to meet a national emergency, was a year old when he began to reign (1 Samuel 13:1 ); but it is impossible to say whether the original text attributed to him twenty, thirty, forty, or any other particular number of years. Nonsense is unfortunately more serious in the original language than in a version: we may pass easily from nonsense in the LXX [Note: Septuagint.] to the actual original consonants of the Hebrew text, which merely require, when thus recovered, to be correctly interpreted; but if the Hebrew letters themselves yield nonsense, we are reduced to guessing, and frequently with little hope of guessing right.
30. The preceding paragraphs should have suggested the justification for conjectural emendation in the textual criticism of the OT, and at the same time they should have indicated its limitations. As against a conjectural emendation, it is in no way to the point to urge that the Hebrew text and all the versions are against it; for the agreement of the Hebrew text and the versions merely establishes the text as it was current about, let us say, b.c. 300. The principle of conjecture is justified by the centuries of transmission that the Hebrew text had passed through before that date. It may be worth while to notice also the degree of truth and the measure of misunderstanding involved in another common objection to conjectural emendations. Tacitly or openly it takes this form: Critics offer different emendations of the same passage; not all of these can be right; therefore the Hebrew text is not to be questioned. The real conclusion is rather this, The fact that several scholars have questioned the text renders the presence of corruption probable, that they differ in their emendations shows that the restoration of the original text is uncertain. The idiosyncrasy of a single scholar may lead him to emend the text unnecessarily: the larger the number who feel compelled to pronounce it unsound, the greater the probability that it is unsound, however difficult or uncertain it may be to pass beyond the negative judgment to positive reconstruction of the text.
31. Evidence of parallel texts within the OT . We have now to consider in what ways beyond those indicated in § 28 the Hebrew text, taken by itself, gives indication of the presence of corruptions, or, on the other hand, of having been accurately preserved, and how it is to be used in order to approximate most closely to the original text, and through it to the original intention of the authors of the several books.
Of most importance, so far as it is available, is the evidence of double texts within the OT. There are certain passages that occur twice over in the OT: e.g. Psalms 18:1-50 is found also in 2 Samuel 22:1-51; Psalms 14:1-7 recurs as Psalms 53:1-6; 2 Kings 18:13 to 2 Kings 20:19 is (for the most part) repeated in Isaiah 36:1-22; Isaiah 37:1-38; Isaiah 38:1-22; Isaiah 39:1-8; 2 Kings 24:18 to 2 Kings 25:21 and 2 Kings 25:27-30 in Jeremiah 52:1-34 , and large parts of Samuel and Kings are incorporated in Chronicles. The variations between these parallel texts are of two kinds: some are due to the editor who incorporates in his own the matter common to his work and the earlier work from which he derives it; for example, in drawing on the Books of Samuel and Kings, the Chronicler often abbreviates, expands, or modifies the passages he borrows, with a view to adapting them to his special purpose; or, again, the editor who included the 14th Psalm in the collection in which Psalms 53:1-6 stands, substituted ‘God’ for ‘Jahweh’ (Psalms, § 2 (2)). With these changes, which it is the province of higher criticism to consider and explain, we are not here concerned. But the second type of variations is due to accidents of transmission, and not infrequently what is evidently the earlier reading is preserved in the later work; ‘and the explanation is very simple: the earlier books were more read and copied: and the more a book is used, the worse is its text’ (Benzinger). In certain cases there is room for doubt as to the type to which particular variations belong, so, for example, in several variations as between 2 Kings 18:1-37; 2 Kings 19:1-37; 2 Kings 20:1-21; 2 Kings 21:1-26 and Is 36 39. As an illustration of the nature and extent of variations between two parallel texts of the OT, we may rather more fully analyze the variations in Psalms 18:1-50 and 2 Samuel 22:1-51 . In a few cases the Greek version of both passages agrees with the Hebrew of one, and here the presumption is that the Hebrew text of the other passage has suffered corruption after the date of the Greek version; but in the majority of cases in which the Hebrew variations can be represented in Greek, the Greek version of Psalms 18:1-50 agrees with the Hebrew text of the Psalm, and the Greek version of 2 Samuel 22:1-51 with the Hebrew text of that passage. In these instances the presumption is that the variation had arisen before the date of the Greek version. There are in all more than 80 variations. Of these just over 20 are cases of vowel letters (§ 6 ) present in the one text, and absent from the other; in the great majority of instances it is the Psalm that has the vowel letters, and 2 Samuel 22:1-51 that lacks them.
Among the remaining variations are cases of the following kinds: (1) Omissions or additions: Psalms 18:1 is absent from 2Sam., so also is Psalms 18:35 b; on the other hand, 2 Samuel 22:8 c is absent from the Psalm. In about a dozen other instances single words present in one text are absent from the other; (2) in two or three cases a word has been lost through the substitution for it of a word-repeated in a parallel or neighbouring line: so ‘billows’ in Psalms 18:4 has accidentally given place to ‘cords’ from Psalms 18:5 (cf. 2Sam.); (3) the variations from Psalms 18:11 b, Psalms 18:42 b in 2 Samuel 22:12; 2 Samuel 22:43 are due to the confusion of similar letters; (4) Psalms 18:28; Psalms 18:31 differs from 2Sam. in respect of the Divine name used (in Psalms 18:31 the Ps. has Eloah, 2Sam. El); (5) inversion of words (not shewn in EV [Note: English Version.] ), Psalms 18:49; there are also cases of inversion of letters; (6) use of different synonyms, Psalms 18:48 . The variation of Psalms 18:14 b from 2 Samuel 22:15 b is more complicated, and the significance of several of the variations is clear only in the Hebrew.
32. Evidence of mutilated literary forms . (1). Acrostics . Thus the comparison of parallel texts furnishes one line of evidence of the way in which the Hebrew text had suffered in transmission before the date of the Greek version. Another proof may be found in the mutilated form in which certain fixed literary forms survive in the present Hebrew text. Most conclusive is the case of the acrostic poems (see Acrostic). At times two considerations converge to prove a particular passage corrupt. For example, the early part of Nah I consists of a mutilated acrostic: in the middle of v. 4 a word beginning with D [Note: Deuteronomist.] should occur; instead, the word × MLL beginning with × is found; but this word × MLL occurs again in the parallel line; in the light of Psalms 18:4 (see previous §, instance 2) it is probable that × MLL in the first has been accidentally substituted for a parallel word which began with D [Note: Deuteronomist.] .
33. (2) Rhythm and strophe . It is possible that further study of the laws of Hebrew rhythm or metre may give us a valuable instrument for the detection of corruption; much has already been attempted in this way, and in some cases already with results of considerable probability. Similarly, in some cases the strophic division of poems admits of conclusions that are again, if not certain, yet probable. Thus in Isaiah 9:8 to Isaiah 10:4 and Isaiah 5:26-29 we have a poem in five strophes marked off from one another by a refrain (Isaiah [Book of], p. 390 a ): in the present text the first strophe consists of 13, the second of 14, the third of 14, the fourth of 14, and the fifth of 15 lines; the probability is that originally each strophe was exactly equal, and that the first strophe has lost a line, and that the fifth has been enlarged by the interpolation of a line.
34. Limited extent of corruption of text of OT . The considerations adduced in the two preceding paragraphs have a double edge. They show, it is true, that the Hebrew text has in places suffered considerably; but they also indicate certain limits within which corruption has taken place, or, to state it otherwise, the degree of integrity which the transmitted text has preserved. If in the ways just indicated we can detect the loss or intrusion of lines or words, or the substitution of one word for another, we can elsewhere claim a strong presumption in favour of a poem having preserved its original length and structure. For example, the majority of the acrostics have come down to us with little or no mutilation that affects their length or the recurrence at the right place of the acrostic letters. Similarly the very possibility of determining rhythm must rest on a considerable amount of the text having reached us free from far-reaching corruption. A further consideration of a different kind may be found in the fact that a large number of proper names (which are peculiarly exposed to transmissional corruption) as handed down in the Hebrew text have been paralleled in ancient material brought to light by modern discovery. In many cases it is beyond question that names have suffered in the course of transmission; but the correct transmission of rare, and in some cases strange, names is significant.
35. Secondary nature of vowel letters: bearing on textual criticism . So long as we deal with parallel texts, we are not brought face to face with the question of how to deal with a Hebrew text resting on a single authority. Yet the great bulk of the OT is of this class. How, then, is it to be dealt with, especially when there is no control over it to be obtained from fixed literary forms? The first duty of sound criticism is to disregard, or at least to suspect, all vowel letters (see § 6 ). We cannot, indeed, assert positively that the original writers made no use of these letters, for we find them employed in certain cases in early inscriptions (Moabite stone, Siloam inscription); but in view of the evidence of the parallel texts of the Hebrew Bible, of the LXX [Note: Septuagint.] , and of Rabbinic references, it is certain that in a large number of cases these vowel letters have been added in the course of transmission. The consequence is that we cannot claim any particular vowel letter for the original author; he may have used it, he may not: particularly in the case of earlier writers, the latter alternative is as a rule the more probable. In other important respects the form of the present Hebrew consonantal text differs from what there is reason to believe was its earlier form.
36. Similarity of certain letters a source of confusion . We have seen above (§ 17 ) that the alphabet in which existing Hebrew MSS are written differs widely from that in use at the time when the OT was written; the letter yod , proverbially the smallest ( Matthew 5:15 ) in the alphabet in use since the Christian era, was one of the larger letters of the earlier script. It is necessary in doubtful passages to picture the text as written in this earlier script, and to consider the prob
These files are public domain.
Text Courtesy of BibleSupport.com. Used by Permission.
Hastings, James. Entry for 'Text, Versions, and Languages of Ot'. Hastings' Dictionary of the Bible. https://www.studylight.org/dictionaries/eng/hdb/t/text-versions-and-languages-of-ot.html. 1909.
